Morphology as a risk factor for the malignant potential of T2 colorectal cancer

Risk of Lymph Node Metastasis in T2 Colon Cancer: A Nationwide Population-Based Cohort Study

BACKGROUND

Similar to T1 colon cancer (CC), risk stratification may guide T2 CC treatment and reduce unnecessary major surgery. In this study, prediction models were developed that could identify T2 CC patients with a lower risk of lymph node metastasis (LNM) for whom (intensive) follow-up after local treatment could be considered.

METHODS

A nationwide cohort study was performed involving pT2 CC patients who underwent surgery between 2012 and 2020, using data from the Dutch ColoRectal Audit, which were linked to the Nationwide Pathology Databank. Four machine learning models were evaluated to predict LNM.

RESULTS

LNMs were found in 1877/9803 patients (19.1%). Independent risk factors included (younger) age (odds ratio [OR] 0.98, 95% confidence interval [CI] 0.979-0.990), left-sided CC (OR 1.5, 95% CI 1.4-1.7), poor differentiation (OR 1.7, 95% CI 1.4-2.2), and lymphovascular invasion (LVI; OR 4.1, 95% CI 3.6-4.7). A deficient mismatch repair (MMR) status significantly lowered the risk of LNM (OR 0.3, 95% CI 0.2-0.5). The general linear model demonstrated the highest prediction accuracy, achieving area under the receiver operating characteristic curves of 0.67 and 0.68, with good calibration. In the absence of risk factors, elderly patients (≥74 years of age) had a predicted risk of LNM of 10.7%, yet up to 30% experienced postoperative complications, with mortality rates reaching up to 3.5%. Patients with a deficient MMR status had a predicted risk of LNM of 6.1% if LVI was absent and the tumor was well-differentiated.

CONCLUSIONS

The risk of LNM should be weighed against surgical risks. The findings of this study will enable clinicians to make more deliberate considerations about these competing risks before making a shared decision.

Linear endo-ultrasonographic signs of muscularis propria invasion in early rectal cancer

BACKGROUND AND STUDY AIM

Local resection of early rectal cancer is being increasingly used. With invasion of the muscularis propria layer of the rectal wall, the risk of lymph node metastasis becomes too high to consider this the optimal oncological treatment. Therefore, a diagnosis of muscular invasion is important before attempting local resection; however, endoscopic and magnetic resonance imaging (MRI) images have limitations, such as overstaging (26-31%). We investigated the potential of linear endoscopic ultrasound (L-EUS) in the diagnosis of muscularis propria invasion.

PATIENTS AND METHODS

The study consisted of a development phase, in which linear (L)- EUS features, associated with muscular wall invasion were searched and tested, and a validation phase, during which 30 representative videos were assessed by the author F.t.B. and four experienced endosonographists without experience in rectal L-EUS.

RESULTS

The development cohort consisted of 91 patients (2019-2023). Overall, six EUS features were found to be significantly associated with muscular wall invasion: tornado sign, blob sign, massive connection, layer split, extramural deposit, and, most importantly impaired shiftability between the lesion and muscularis propria layer. During the development phase, these findings demonstrated excellent diagnostic features (sensitivity, 94.4%; specificity, 97.9%; and overstaging, 4%). In the validation phase, the sensitivity, specificity, and overstaging by F.t.B. were 88%, 85%, and 12%, respectively. Among the four inexperienced reviewers, the percentages were 65%-71%, 46%-54%, and 33%-39%, respectively. When considering the 27 videos that were considered easy or moderately difficult to assess, only 55% were correctly interpreted by the inexperienced reviewers.

CONCLUSIONS

Linear endoscopic ultrasonography may be a valuable tool for the assessment of ingrowth into the muscularis propria in supposedly early rectal cancer, especially using its dynamic potential to assess fixation to the muscular wall by moving the lesion. However, training will be required to achieve satisfactory results.

Deep learning for segmentation of colorectal carcinomas on endoscopic ultrasound

BACKGROUND

Bowel-preserving local resection of early rectal cancer is less successful if the tumor infiltrates the muscularis propria as opposed to submucosal infiltration only. Magnetic resonance imaging currently lacks the spatial resolution to provide a reliable estimation of the infiltration depth. Endoscopic ultrasound (EUS) has better resolution, but its interpretation is investigator dependent. We hypothesize that automated image segmentation of EUS could be a way to standardize EUS interpretation.

METHODS

EUS media and outcome data were collected prospectively. Based on 373 expert manual segmentations, a convolutional neural network was developed to perform segmentation of the submucosa, muscularis propria, and tumors. The mean surface distance (MSD), maximal distance between segmentations (Hausdorff distance; HDD), and overlap (Dice similarity index; DSI) were calculated.

RESULTS

The median MSD and HDD values were 3.2 and 17.7 pixels for the tumor, 3.4 and 24.7 pixels for the submucosa, and 2.6 and 20.0 pixels for the muscularis propria, respectively. The median DSI values for the tumor, submucosa, and muscularis propria were 0.82, 0.57, and 0.59, respectively. These values reflect good agreement between manual and deep learning segmentation.

CONCLUSIONS

This study found encouraging results of using automated analysis of EUS images of early rectal cancer, supporting further exploration in clinical practice.

Molecular and clinicopathological differences between depressed and protruded T2 colorectal cancer

Background

Colorectal cancer (CRC) can be classified into four consensus molecular subtypes (CMS) according to genomic aberrations and gene expression profiles. CMS is expected to be useful in predicting prognosis and selecting chemotherapy regimens. However, there are still no reports on the relationship between the morphology and CMS.

Methods

This retrospective study included 55 subjects with T2 CRC undergoing surgical resection, of whom 30 had the depressed type and 25 the protruded type. In the classification of the CMS, we first defined cases with deficient mismatch repair as CMS1. And then, CMS2/3 and CMS4 were classified using an online classifier developed by Trinh et al. The staining intensity of CDX2, HTR2B, FRMD6, ZEB1, and KER and the percentage contents of CDX2, FRMD6, and KER are input into the classifier to obtain automatic output classifying the specimen as CMS2/3 or CMS4.

Results

According to the results yielded by the online classifier, of the 30 depressed-type cases, 15 (50%) were classified as CMS2/3 and 15 (50%) as CMS4. Of the 25 protruded-type cases, 3 (12%) were classified as CMS1 and 22 (88%) as CMS2/3. All of the T2 CRCs classified as CMS4 were depressed CRCs. More malignant pathological findings such as lymphatic invasion were associated with the depressed rather than protruded T2 CRC cases.

Conclusions

Depressed-type T2 CRC had a significant association with CMS4, showing more malignant pathological findings such as lymphatic invasion than the protruded-type, which could explain the reported association between CMS4 CRC and poor prognosis.

Novel "resect and analysis" approach for T2 colorectal cancer with use of artificial intelligence

BACKGROUND AND AIMS

Because of a lack of reliable preoperative prediction of lymph node involvement in early-stage T2 colorectal cancer (CRC), surgical resection is the current standard treatment. This leads to overtreatment because only 25% of T2 CRC patients turn out to have lymph node metastasis (LNM). We assessed a novel artificial intelligence (AI) system to predict LNM in T2 CRC to ascertain patients who can be safely treated with less-invasive endoscopic resection such as endoscopic full-thickness resection and do not need surgery.

METHODS

We included 511 consecutive patients who had surgical resection with T2 CRC from 2001 to 2016; 411 patients (2001-2014) were used as a training set for the random forest-based AI prediction tool, and 100 patients (2014-2016) were used to validate the AI tool performance. The AI algorithm included 8 clinicopathologic variables (patient age and sex, tumor size and location, lymphatic invasion, vascular invasion, histologic differentiation, and serum carcinoembryonic antigen level) and predicted the likelihood of LNM by receiver-operating characteristics using area under the curve (AUC) estimates.

RESULTS

Rates of LNM in the training and validation datasets were 26% (106/411) and 28% (28/100), respectively. The AUC of the AI algorithm for the validation cohort was .93. With 96% sensitivity (95% confidence interval, 90%-99%), specificity was 88% (95% confidence interval, 80%-94%). In this case, 64% of patients could avoid surgery, whereas 1.6% of patients with LNM would lose a chance to receive surgery.

CONCLUSIONS

Our proposed AI prediction model has a potential to reduce unnecessary surgery for patients with T2 CRC with very little risk. (Clinical trial registration number: UMIN 000038257.).

Evaluation of radical surgical treatment in the management of 58 locally advanced rectal neuroendocrine neoplasms, one multicenter retrospective study

BACKGROUND

Locally advanced rectal neuroendocrine neoplasms (NENs) are rare, and the therapeutic effects of surgery in improving the prognosis have been questioned in previous reports.

MATERIALS AND METHODS

The research included 58 consecutive patients with locally advanced rectal NENs from three Chinese medical centers between 2000 and 2020. All have received radical surgical treatment. The clinicopathological and survival data were collected. Kaplan-Meier methods and a Cox proportional hazards regression model were used to evaluate the prognosis and identify independent prognostic factors.

RESULTS

All patients were followed up for a median period of 36 (2-125) months. Of the 58 patients, 13 (22.4%) had G1 neuroendocrine tumors (NETs), 15 (25.9%) had G2 NETs, 6 (10.3%) had G3 NETs, and the remaining 24 (41.4%) patients had G3 neuroendocrine carcinomas (NECs). The 1-year and 3-year disease-free survival (DFS) rates were 64.5% and 48.8%, respectively. The 1-year and 3-year overall survival (OS) rates were 90.5% and 75.4%, respectively. Univariate analysis demonstrated that tumor differentiation (p = 0.002), gross morphology (p = 0.009), T stage (p = 0.024), and extramural vascular invasion (p = 0.009) were associated with the OS. The subsequent multivariate analysis confirmed that tumor differentiation [hazard ratio (HR) = 6.002, 95% confidence interval (CI): 1.210-29.767, p = 0.028] and gross morphology (HR = 3.438, 95% CI: 1.038-11.382, p = 0.043) were independent prognostic factors affecting the clinical outcomes.

CONCLUSIONS

Rectal NENs are a heterogeneous group of diseases. The survival benefits obtained from surgery vary widely based on the tumor clinicopathological features. Patients with G3 NECs and ulcerative mass are at high risks of poor prognosis.

Tumor Macroscopic Morphology Is an Important Prognostic Factor in Predicting Chemotherapeutic Efficacy and Clinical Outcomes of Patients With Colorectal Neuroendocrine Neoplasms, One Multicenter Retrospective Cohort Study

BACKGROUND AND AIMS

Locally advanced and metastatic colorectal neuroendocrine neoplasm (NEN) is a rare disease with a dismal prognosis. We aimed to explore the value of the macroscopic morphology of NENs in the management of TNM stage II-IV colorectal NENs, which has not been fully elucidated in previous reports.

METHODS

We retrospectively enrolled 125 eligible patients with TNM stage II-IV colorectal NENs who were diagnosed between 2000 and 2020 from three Chinese hospitals. All were categorized into either protruding or ulcerative NEN groups through endoscopic evaluation of their macroscopic morphology. Clinicopathological data were collected and compared between the two groups. Survival analysis was performed to assess the survival outcomes between the two groups.

RESULTS

A total of 77 and 48 patients had protruding and ulcerative NENs, respectively. Patients with ulcerative NENs had a larger median tumor size (P<0.001) and higher median Ki-67 index (P<0.001), and a larger proportion of these patients had grade G3 disease (P=0.001) and poorly differentiated neoplasms (P=0.001), as well as higher frequencies of T3 and T4 tumors (P=0.006) than patients with protruding NENs. In addition, patients with ulcerative NENs showed a much lower response to first-line chemotherapy [50% (95% CI: 27.3% - 72.7%) versus 20% (95% CI: 3.1% - 36.9%), P=0.03] and a worse 3-year progression-free survival (PFS) rate [19.7% (95% CI: 7.2% - 32.2%) versus 49.5% (95% CI: 37.5% - 61.5%), P=0.001] and 3-year overall survival (OS) rate [30.7% (95% CI: 15.6% - 45.8%) versus 76.9% (95% CI: 66.5% - 87.3%), P<0.001] than those with protruding NENs. The multivariate analysis results indicated that the macroscopic shape of NENs was an independent prognostic factor affecting both PFS (HR = 1.760, 95% CI: 1.024 - 3.026, P = 0.04) and OS (HR = 2.280, 95% CI: 1.123 - 4.628, P = 0.02).

CONCLUSIONS

Ulcerative NENs were more malignant and chemotherapy resistant than protruding NENs. Tumor macroscopic morphology is a valuable prognostic factor for stage II-IV colorectal NENs.